Abstract: Research-practice-partnerships (RPPs) have arisen as a potentially powerful mechanism for school improvement; however, there is little work how to evaluate RPPs. This study investigates how four RPPs are addressing impact by (a) document analysis of metrics (N=123) being used to assess partnerships, and (b) interviews exploring how network leads (N=11) and policymakers (N=3) conceptualize partnerships and their impact on the frontlines. Findings suggest that while metrics being used provide a necessary baseline for the number and types of partnerships, more robust methods are needed capture the quality of interactions and to strategically inform network development. The discussion advocates for network improvement through sharing cases of failures (alongside exemplary cases) to maximize learning, and for the use of developmental evaluation to explore the impacts of RPPs.